Methods Behind the Madness: Why Only Count the Most Birds Seen at Once?
You’ve just sat down in your living room under a cozy blanket with a steaming mug of fresh coffee to begin your FeederWatch count. You watch for an hour as chickadees race back and forth to your feeder from deep inside spruce trees just outside of view. You know there are probably several birds darting around in the branches, but they only come into view one at a time. How do you record these multiple visits in your count?
While it’s tempting to add up every visit by the chickadees, the FeederWatch protocol calls for reporting only the highest number of each species seen simultaneously. Why is that? It is because the protocol is designed to prevent you from counting the same bird more than once. If you counted every visit to your feeders, you would report more individuals than you actually have, because some birds—especially species like chickadees—visit multiple times.
Let’s imagine an example. In the morning, you watch your feeders and only see one chickadee at a time, even though you see about 20 different visits over the hour. You report “1” for your chickadee count that morning, because you only ever saw one bird at a time. Later in the day you sit down again (maybe with some afternoon tea), and you spot four chickadees all at once! Now you can increase your count for chickadees to “4.” You don’t add the morning and afternoon counts together because the chickadee you saw in the morning might be among those visiting in the afternoon. The next day you watch your count site for the second day of your two-day count, and you see three chickadees at the same time. Should you add these three to the four you saw the day before? No, because those three might be the same birds you saw the day before. You should report four chickadees when submitting your data because you can only be certain that you have at least four chickadees in your yard.
By reporting only the highest number of each species seen at the same time, is it possible that you are missing some individuals? Yes, of course, but that is OK. The most important thing is that everyone in FeederWatch counts the same way, even if the counts aren’t recording the exact number of birds in your yard. Knowing the exact number is impossible unless birds are uniquely marked and do not move in and out of your yard—which is extremely unlikely! FeederWatch counts are a sample, or estimate, of what is in your yard. If everyone samples in the same way, then counts across species, space, and time are comparable to one another, allowing us to build a continentwide, three-decade-long dataset of bird abundance.
The need to count all species in the same way is also why you should not count males and females separately if they visit your site at different times. You may know that you have a male and a female cardinal in your yard, but unless you saw both at the same time, you should report only one cardinal. This may feel wrong, but trust us, it is OK! The most important thing is to sample all species in the same way, so that we can estimate population change
18 comments on “Methods Behind the Madness: Why Only Count the Most Birds Seen at Once?”
“Estimate pop’n change” is NOT the same as COUNTING birds. Statisical modeling is NOT birdwatching. So you get no data from me, except on special count days. How much data have you lost by being persnickity? Identifyable individuals (banded or sexed, whatever) should not be disregarded in a count because you want to play Mathamatics.
Hi Bahb, Statistical modeling is based on data that is collected in a standard way (in this case, the standard is the FeederWatch protocol). The inferences we can make from the modeling portion of data analysis won’t be accurate if different people count their birds in different ways. For example, we would not be able to compare data for Northern Cardinals, where both male and female cardinals were counted separately, to Carolina Chickadees (males and females look the same) because the model would be biased towards cardinals rather than comparing the two from the same baseline. So, in this example, we would not be able to accurately say whether a yard has more cardinals than chickadees, because the cardinals were counted differently than chickadees. So, to make accurate assumptions, all birds must be counted the same way according to the FeederWatch protocol.
Please note that the days that are chosen for counts should be unbiased as well (i.e. it’s important not to start a FeederWatch count just because you see an exciting bird). The real value in FeederWatch counts comes from unbiased count days and everyday birds. Please feel free to check out our previous blog post about biased counts for more details and an example showing how it can affect data output.
If you are interested in reporting every individual bird that you see, I recommend checking out eBird, another project here at the Cornell Lab of Ornithology.
Would that ebird worked for me. I started listing birds in Spring of 2021, but ebird crashed after 20 sightings. I keep my own list on paper.
Trying to start our feederwatch program but need some clarification. The instructions say count on 2 consecutive days for about an hour each time. The same hour each day is required. At our feeder Chickadees, Finches, and Juncos all come randomly at all hours all day. How do you count these birds. once every 5 minutes of the hour or as many that show up over the hour?
Hi Jere, Your two-day count should indeed be conducted over two consecutive days, but you can count as much or as little as you want, and the time you count each day does not have to match. You could count for 5 minutes in the morning one day and 6 hours straight on the next day, or could even count on and off all day as you do chores around the house. Just make sure that when you enter your data, you are accurately reporting how much time you spent actually watching your count site (e.g., if you watched for about 10 minutes each hour for 6 hours, only report those 60 minutes you spent watching) and you’ll have a chance to report whether you watched in the AM, PM, or both for each day. You can read our instructions online on our website here for more detail.
Just a newbie on this and your instructions are very helpful! Thank you! Especially liked the “birdfeeder picture” and notes below. Excited to be a part of this!
We don’t see , for the last 3 months, any chickadees anymore, what’s going on?
We live in harmony ( close to Truro Nova Scotia)
Normally plenty of chickadees to our feeder.
Hi George, Although it’s impossible for us to know the cause of each specific increase and decline of birds, there are several common causes for bird population fluctuations:
* Habitat changes frequently affect bird numbers. If there has been any change in your neighborhood–such as trees being cut down, new houses being built, or different crops being planted on nearby fields–that could be the reason you are seeing more or less birds.
* Natural food supplies–such as pine cones, berries, seeds, and insects–fluctuate from year to year, causing birds to shift ranges to take advantage of food surpluses or to compensate for food shortages.
* Weather fluctuations often cause birds to shift ranges, especially in winter.
*Predators such as hawks or domestic cats sometimes move into an area and cause birds to move elsewhere.
You can see a summary of FeederWatch reports so far this year for Nova Scotia here – note that the season just started so many folks haven’t reported their counts yet, but you can keep an eye on this list to keep track of what species other people are seeing.
We struggle with the feral cats. Birds to not congregate around our feeder due to the fact that cats lie in wait for them. No matter where we put the feeders the cats climb the trees, wait under the feeder hutch, even wait in the bushes where the birds nest in the summer. I’m at my wits end. I wanted to participate in the last three counts but even the crows won’t land in the yard. The cats are damaging our native bird population. It’s quite frustrating. Is there anything I can do without putting the birds at risk?
Hello Velynda. We’ve heard that putting a short chicken-wire fence around a ground feeding area below your feeders will keep cats out of that area, but we haven’t tested it. Apparently cats are reluctant to jump over chicken wire. The fence needs to be high enough to make jumping over it hard but short enough for you to easily step over it to access your feeders. Another option we haven’t tried is to spray the area around your feeders with a cat deterrent. There are products available to keep cats from eating house plants, for example, but we don’t know how well they work outside, and you might need to spray each time it rains. If you try either of these options, let us know how well they work for you.
Aside from that, you can hang feeders out of reach of cats and rake the ground beneath the feeders frequently to discourage birds from gathering there. Or you can set up a catch basin under the feeders to catch fallen seed and cover it with a screen that has holes big enough for seeds to fall through but too small for birds or rodents to get into. This option has the added benefit of deterring small rodents as well, but you’ll need to empty and clean it frequently to keep droppings from building up and clogging the holes, and you may need to take it in at night to keep larger mammals, like raccoons, from tampering with it.
I am posting here off topic as others have since it is the most recent post. I am looking for better guidance about the area around the feeders that I should count. For example, the feeder attracts birds, and the birds occasionally attract a Cooper’s hawk. But the hawk doesn’t land on my feeder. If he’s flying overhead or up in a tree 75 feet away, do I count him? Likewise just random birds in the general vicinity, like crows. Thanks.
Hi Heather, you can read more about how to count birds here. If you have more questions about participation, please email feederwatch@cornell.edu. These comments are not regularly monitored.
Ok it just indicates that hawks that are nearby and seemingly attracted to the birds at the feeder can be counted. I guess I can just make a decision about crows and raptors who are at a certain distance like 100 feet or 200 feet, and count them. The instructions don’t get more specific than that. thanks for the contact email.
Call your county animal control. They can at least trap them and nuder them. I have used a machine gun type sprinkler connected to a remote hose connection that I can turn on un seen. They stay away now
Cats aren’t our problem, squirrels are. The small Red Squirrels of South Central Alaska, west Kenai Peninsula, near Cook Inlet, rural.) Been here14 years. Had to glue thin sheet-metal around each access to one of those tall square sunflower feeders, to stop them from chewing their way in. Roof overhangs 18″; feeder under it at window (too close for birds to get up enough speed to hurt themselves). The Squirrels can climb our T-111 siding; that’s why it’s moved to window.
Red-Breasted Nut Hatches often take out up to three seeds before finding one they “like”, tossing rejected seeds on porch deck. On the deck, the LBBs have to compete with the squirrels, but squirrels nest some distance away, so not that many over space of an hour. It appears some birds use the porch deck when competition at feeders is too much.
Feeders have to hang above reach of mice. Saw three in feeder, another climbing, when pagoda was on a stump. The cat we brought from former home solved that – and he was a very bad bird catcher. The chickadees were good at sounding the alarm, too. The red-backed voles seem to stay off deck. Cat died, haven’t “replaced” him. Our dogs chased squirrels, did a good job. Never caught one that we saw, or evidence.
Other feeder wooden pagoda 14″ across, not completely under overhang. The red-breasted nuthatches use its roof to peck open sunflower seeds. Stopped using it for a summer to stop squirrel expectations. (Also: When you live with bears, you don’t feed them sunflowers or suet – duh – but garden fenced high & strong for moose.) I add Nyger (Russian thistle) to pagoda. Never found a mixed-seeds variety that is popular. Those hard, darkish red, round seeds stay in feeder until I clean it. Corn pieces not popular, but hard to find suet without it. Have to live-trap squirrels with unsalted peanuts-in-the-shell without attracting birds that eat those.
At right time of year, the quart-sized sunflower feeder empties in no more than 2 days, suet goes quick.
2 big lilac bushes within 5 feet of feeders give safe perches, protect from winged predators.
Chicken wire, stapled loosely on fence, helps keep out porcupines, who damage/kill berry bushes and trees that birds like for food and perches. Squirrels as small as ours might not be possible to keep out.
One summer I was getting hummingbirds so often that I was practically tripping over them. I knew I wasn’t getting the dozens it appeared so I made a query and was told to count (for a period of time, I believe) and divide by 3. It worked for me and the number seemed reasonable.
While I can’t make the logic of your protocol work for me, it’s your protocol. You don’t have to justify it to me but to the scientists/ornithologists you report to. So I’ll happily sit in front of my window and collect bird sightings, new birds every few days, coffee in hand and a cookie nearby. Then in the evening I’ll send the numbers to you and you’ll make sense of them and we’ll all be happy. Happy, happy, happy!
I watched the session for training people for the great back yard birdcount and they can count 2 birds if they see a male and female separately whereas feederwatch cannot. Why this difference?
Over the last many years, our cardinal population has increased. It seems that the young return with the parents. As we have added more native plants to the yard, the goldfinches have dramatically increased. We notice new couples in the yard : the gray catbirds, the red-bellied woodpeckers, the titmice. However, our chickadee population seems to remain a constant. We have a pair, and it seems to stay at that! I can only ever see one during my feederwatches. As soon as I close my laptop and head out in the yard, there are the two of them in an osier, but never during my counts, haha. Still, my question : where are the young? Are chickadees very territorial? Do they have to move a certain distance from the parents? Thanks for your time.